Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Customer Journey as a Service (CJaaS)Technology Architecture, Cloud Platform & Open API
Nagendra C L, Chief Architect Feb 2021
© 2021 Cisco and/or its aliates. All rights reserved.
üCustomer Journey Platform
Flow
Graph
Insight
Analytics
Builder
Creation
Design
Evolution
Management
Profile
Visualization
Mapping
Orchestration
</>
Select All
White paperCisco public
Introducing
Customer Journey as a ServiceCPaaS Open APIAvailable ~Feb’21/Experimental
Soon to be built with at
XM 2.0
For CCaaS (Contact Center)CJaaS | Ensemble Routing | VOC Outcome Framework
By Webex Experience Management
Customer Journey as a Service(CJaaS)Technology Architecture, Cloud Platform & Open API
Nagendra C L, Chief Architect / Cisco Webex Experience Management Aug 2020
üCustomer Journey
Builder
Visualization
Creation
Graph
Platform
Evolution
Orchestration
Design
Insight
Analytics
Profile
Flow
Mapping
Management
</>
Select All
White paperCisco public
Introducing
CCBU Experimental
Incubation Series
CCBU Experimental
Incubation Series
Introduction to JourneysWhat is a Customer Journey ?
CRM Centric - Outbound customer journeys are built by hand (aka “orchestrated”) to enable slice/dice and adap�ve messaging applied to a workflow over �me. Typically, solu�ons include "Marke�ng Automa�on" of yesteryears that eventually evolved to offer “Journey Builders of Today”.
A customer journey is a visual representa�on of customer interac�ons, experiences and ac�vi�es - captured and/or orchestrated across channels and over �me.
Regardless of the fields, journeys are built from sequences of events that an individual (”the customer”) may take while interac�ng with a brand, this could o�en be a poten�al customer who may be looking to buy a product or service or a paying customer looking for delivery of a service.
Channels: Primarily driven off CRM, via Email, SMS, Push or Social Media.Solving for: Programmable Messaging for Sell, Cross-sell, UpSell, MQL, etc.
Each interac�on at every touchpoint leaves an event trail, that could be anything from browsing a page (tracked via digital beacon), the click of an ad (clickstream), reading of an email (pixel tracker), adding an item to a cart (pushed api event), talking to a agent or even just few moments of pause to look at a display while at a physical store (captured by a IoT beacon) – each and every interac�on contains �ny bits of informa�on that is stored (on a CDP) as an series of events leading up to that very moment of truth. Knowing this sequence of events, adds to intelligence and insight that allows a company to personalize an interac�on (MarTech), to op�mize for revenue (Adtech), or simply to empathize while delivering an great experience (Contact Center).
Customer Journeys are used in several fields, but each field implements a different flavor, depending on the circumstances. Regardless of the terminology, each solve for different objec�ves with unique analy�cs to enable different outcomes. Generally tools & services built for one segment do not easily translate to solve problems for others. Let us look at the 4 major flavors:
#1 MarTech Journey (CRM Centric)
Messaging
CRM
(Source) Segment
("all contacts from Texas")
Send Email
(”the download oer”)
Landing Page
Not Opened
Click-through
Programmed Wait-Time
(”3days”)
Reengage Follow up
Engagement Split
SMS
Add to MQL Call List
86,400 3,600
2200
1400
3,600
[ Human-in-the-loop ]
White paperCisco public
© 2021 Cisco and/or its aliates. All rights reserved.
CCBU Experimental
Incubation Series
Introduction to Journeys
Web ad display centric journeys are visualiza�ons of single or mul�-site clickstreams (as in unfiltered web clicks/naviga�on), primarily tracking poten�al customers across one or more pages as they move about with goal to op�mize ads, drop-offs and click-throughs. These are typical web analy�cs / op�miza�on systems of yesteryears evolving to offer "Journeys” of Today.
Channels: Primarily driven off cookies, via Web beacon trackers (even mul�site).Solving for: Improve ads, increase click-thoughs, be�er conversions.
Solving for: Improve Service Delivery, Brand Experience
While “experience” could in theory be es�mated, it likely will not be a replacement for NPS/CES as rated by a real human input and today s�ll remains in realm of experimental research.
Voice of Customer (VoC) Centric Experien�al Customer Experience Management (CEM) journeys are primarily visualized around touchpoints (”Aggregates"), driven off user surveys that segment based on KPI metric such as NPS/CES to follow perceived customer experience across touchpoints to op�mize customer service.
Channels: Survey, Intercepts, Ques�onnaires
#2 AdTech Journey (Web Centric)
#3 CEM Journey (VoC KPI Centric)
Landing
Page
Conversion
Defer
Incoming Traffic
Organic
Referral
Paid
Bounce
Loop Page Flow
Booking a Ticket On the Web Checkin At the Gate In-flight Service BaggageŒ � Ž � �
Flight #342
-vs-
Flight #3445:40 AM
21:10 PM
NPS/CES
© 2021 Cisco and/or its aliates. All rights reserved.
[ Human-in-the-loop ]
White paperCisco public
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
Introduction to Journeys
Contact center journeys (inbound) revolve around agents and their channels, that usually track the channels to maximize for efficiency, to lower average handling �me (AHT) and/or improve customer sa�sfac�on.
Agents aware of near term context at the �me of interac�on are be�er placed to field the customer and provide faster resolu�on �mes.
#4 Contact Center Journey ( Agent Centric )
ChannelsWebsite
Visit Auto Loan EV Financing Quote Request to Callback
Email OpenedEmail Interaction
Outbound Agent Interaction
Inbound CallApplied
5 hr ago 5 hr ago 5 hr ago 4 hr ago 1 hr ago
10:52s AHT [ Resolution: Quote Emailed ]Customer
45 min ago30 min ago
10 m ago
2 min ago
Agent 12XX
Agent 14XX
Interactions
Outbound only contact center journeys also may include outbound offers to connect with agents, using journeys built for intercepts that display content to help with maximizing u�liza�on of available agent minutes for revenue.
Solving for: U�liza�on (inbound contact center) or Revenue (outbound opportuni�es)Channels: Web tracking w/ Inbound Voice/Chat/Email or Outbound proac�ve reach out
[ Human-in-the-loop ]
Select JS Trigger Events
CDR Events Pixel Track
è CJ as a Service
White paperCisco public
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
Customer Journey: Feature Transitivity Across Fields
Ÿ AdTech: Not applicable, as Journeys are being observed rather than being designed.
Orchestra�on-Journey Builder:
The challenges of comparing journey feature capabili�es across fields.
Elements of the customer journey differ based on the circumstances. Some fields may build workflow to curate the pathway to purchase while others may cover the product adop�on cycle. Different terminology need to be considered in the context of the field, for instance, below is an example of what a “Journey Builder” looks like across fields:
Ÿ MarTech: Core of any marke�ng workflow orchestra�on is the Journey Builder.
Ÿ VoC/CEM: Across channel survey delivery work flow builder is the Journey Builder.Ÿ Contact Center: IVR Designer & “Ensemble AI Rou�ng” orchestra�on could be Journey Builder.
MarTech/CRM
AdTech/Web
VoC/CEM
Contact Center(<2019)
Product Experience
User/Persona Based
Likely Maybe Unlikely
Orc
hest
rate
d/B
uild
er
(Pre
-D
ene J
ourn
eys
)H
um
an-in
-Loop
Clic
k S
tream
(B
lanke
t/C
ookie
Tra
ckin
g)
Conte
nt
Dis
pla
yO
ers
Em
ail
Mark
eting
Pendo, Amplitude, MixPanel
Dynamics, SalesForce, Adobe EM
Google Analytics, MixPanel, Piwik
Cisco Webex XM (~CloudCherry)
AltoCloud
Various Journey Consultants/PS
Example Segment CohortsCJM
(M
ap)
CJA
(A
naly
tics)
CJL (
Laye
rs)
Journ
ey
Vis
ualiz
ation
Surv
eys
Pro
gre
ssiv
e P
role
sC
DP
Designed For:
Om
ni-
Channel
Customer Journey Capabilities: Confusion Matrix
è Meet CJaaS
White paperCisco public
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
CJaaS: Customer Journey as a Service
Ÿ Low latency
Ÿ Delivery of powerful orchestrated ac�on at any point in journey across channels
CJaaS provides a common journey infrastructure that can easily be consumed as a drop-in SDK Communica�ons Pla�orm as a Service (CPaaS) to enable customer journey capabili�es across fields featuring:
Ÿ Dynamically scalable data sinks from any channelŸ Low cost, Serverless APIs
Ÿ Firehose capable of receiving and processing millions of events per secondŸ Ability to query/pipe/subscribe to data streams for custom analy�cs
Applica�ons can select from a menu of pre-built tools and features to build a customer journey capability that fits their specific customer journey needs:
Ÿ CJaaS is the Journeys capability part of larger CPaaS offering (Product Neutral)Ÿ Cloud Contact Center + CJaaS = Na�ve Cloud Customer Journey for Cloud Contact CenterŸ Enterprise Contact Center + CJaaS = Enterprise Contact Center with Customer Journeys on Cloud.Ÿ Experience Management + CJaaS = XM turbo charged with Customer Journey++
By u�lizing a common infra, a unified journey sink and powerful orchestra�on, companies have the op�on to use any product while following their customers’ journeys everywhere, across channels.
CJaaS also enables companies to mix-and-match customer journeys across on-prem, hybrid and cloud environments thus making it a choice for highly regulated sectors such as Finance and Healthcare. Significantly, “as a service” delivery enables the underlying products to invest in core capabili�es while customer journey becomes a plugged-in capability via CPaaS. By moving Customer Journey infrastructure to a CPaaS model, applica�ons can focus their �me and resources in developing innova�ons in core solu�ons.
White paperCisco public
è Technology Stack & Architecture
CCBU Experimental
Incubation Series
Profiles
Pla�orm Built On
Events
Low-latencyNoSQL datastore
ProfilesPro
gres
sive
Jou
rney
Op
en A
PI
cj.x
m.w
ebex
.co
m/a
pi
OPEN API
REST API
.NET SDK
PHP Package
Java Lib
REST/JSON
CJA
/CJO
Vis
ual
iza�
on
W
eb C
om
po
nen
tsC
J In
ges�
on
Ch
ann
els
Jou
rney
To
uch
po
ints Web
Chat
Contact Center
Data Sink
Serverless Func�on+-AutoScale
Customer Journey As A Service Pla�orm
AP
I M
anag
em
ent
Service Bus / Cloud Tasks
Receive
Queue
Return
The “Tape” WriterServerless Func�on
+-AutoScale
Dequeue
Process
Store
Table / Big Table Storage
Analy�cs & Profile
NoSQL Data Store
Profile BuilderServerless Func�on
+-AutoScale
Consume
Build
Cache
Insight
ETL
Map
Dat
a La
ke
Tahoe Reg
istr
y
Flow Control
Orchestrate
Progressive Decaying
Realtime Prole API
Serverless Decoupled
High Capacity Sink
Rated 1M/s Events
Allow ListEven
ts
© 2021 Cisco and/or its aliates. All rights reserved.
Serverless
WebComponents
+MultiCloud
Queue Listener
Ka�a Consumer
Orc
hes
trat
e
Ac�ons
The “Tape” ReaderServerless Func�on
+-AutoScale
Orchersta�on
Events Engine
Analy�cs Aggrega�onSegment Journeys
Serverless Func�on+-AutoScale
Events Streamer
SSE Stream
Transform
Walk-In Rou�ng
Cloud Routing Table
Discovery Delivery
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
Customer Data Platform: Profiles Built On The Go
Customer Journeys are only as good as the context they capture, and with every journey there is a human customer profile that needs to be known, otherwise the context would be lost.
Context can be built incrementally as the journey progresses, instead of at all once. While CRM may hold the bio essen�als - it remains a sta�c snap shot. Progressive profiles are much more than a sta�c record profile that may be held in the CRM. These profiles are dynamic and living as they capture interests - intent and details over �me.
Context is a significant advantage to both agent and AI models in a contact center se�ng. Consider for instance, an exis�ng checking account customer who browses auto loans online before calling in for a rou�ne service request for banking, this may provide an opportunity for a cross-sell while behind the scenes “ensemble rou�ng” can override the IVR and route directly to connect with an agent who has auto loan quo�ng competency, in-addi�on to general servicing to make serendipity more likely.
Example Profile Tracking:
{
UniqueId: "CUST12XX",
Event: "Prole",
Location: "Online Website",
"Details": {
“City”: “San Francisco”
“Model: 2020
"Vehicle Type": "SUV/EV"
}
}
Member Member
Checking
Member
Checking
+Auto Loans
Spent 1m+ on this section
Unique Individual ID(ex: username, relationship #)
“Any” event on event stream
Originating channel
City on File/CRM Dip
Hovered on multiple SUV/EV only oers
è The “Tape”
White paperCisco public
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
Engagement Tracking Across Channels: The Tape
Every customer journey needs to track a wide range of events across touchpoints over a period of �me to best capture the full range of interac�ons that occur when a customer engages with their products and services. Each interac�on is composed of an event that describes: what occurred – when – by whom.
Interac�ons may take place physically at a loca�on/store/branch or online on a website or in an email. Each event is a granular data point from which a journey is built.
Each event data has informa�on about what happened but does not have the data that triggered the event. For instance, an event no�fies that a customer clicked an offer displayed, and it may have general informa�on about the promos, but it does not have the offer image itself.
Example Interac�on Stream:
Customer Journey as a Service: Inbound Tracking Channels
On-Location On-Web In-Product/App With-Agent
IoT Beacon
With-Email
Digital Beacon Plugin SDK Open Tracker Pull/Push API
In-Store/Branch
IoT Beacons
Drop-in one-line
Javascript
Mobile SDK or a
Rest API Call
Drop-in pixel
tracker image
Push or Pull in
Realtime
+
{
Location: "Mobile App",
"Oer": "15% o rst order",
}
// Event Originating In-App
UniqueId: "CUST12XX",
Event: "Promo Clicked",
Details: {
“City": "New York"
}
Context connected with event
(Ex: SKU in a cart)
Unique Individual ID (ex: username, relationship #)
Any named event
Originating Channel
è Journey Orchestra�on
White paperCisco public
IOT WEB APP EMAIL CLOUDEVENTS
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
Customer Journey Orchestration: The Bridge to Possible
è By Example Customer Journey
A: Customer Journey Orchestra�on Makes it Possible!Q: Can Contact Center increase top-line and improve bo�om line at the same �me?
In recent �mes with much of world socially distancing, most people in customer facing jobs have found it challenging to stay produc�ve and fill their quotas/slots with serviceable customers or leads. This includes insurance agents, personal bankers, real estate agents, fitness trainers & even ma�ress sales staff.
Many people may prefer to meet over video rather than in person at a physical loca�on. To match this need with the supply geographically, companies can leverage their en�re workforce as in-the-moment agents to provide just-in-�me resolu�on or sales in their regions, thereby replica�ng the previous experience of transac�ng in a physical world (branch/loca�on).
White paperCisco public
Website
Quote Step-1 Step-2 Step-3
Contact Center
Claims
Customer Support
100%
65%
15%
20%
10% 15% 25% 75%
50% 25% 15%
35%
Quote Purchase
25%15%
75%
Customer Journey Analytics
Journey Orchestra�on draws inputs from Journey Analy�cs to pro-ac�vely intervene in keeping customers happy with personalized content, real-�me agents, or just-in-�me virtual walk-ins. These work across channels, preven�ng what otherwise may have been a drop-off, lost sale or nega�ve experience.
At more than
and if greater than 30 seconds
then within
Full-Time
Employees
21K
Customer Facing
InStore/Branch
9450 (45%)
Administrative
3990 (19%)
Others
4410 (21%)
Contact Center
1370 (7%)
EXAMPLE STAFFING BREAKOUT (2020)
Journey orchestra�on allows companies to create virtual walk-ins on par with the physical walk-ins experience.
As companies increasingly realize that work is an ac�vity instead of a loca�on, Customer Journey Orchestra�on (CJO) empowers workforces to remain agile, nimble and service their customers remotely from any loca�on, while also staying relevant and providing meaningful digital connec�ons.
Work is a Ac�vity, Instead of Loca�on
Journey Orchestrated Actions
Action #1: Journey Drop-o Interception On
OnJourney
Step-3
Wait Time
Action
Walkin Oer
Agent OerContent
+
Drop o
Distance
5 Miles
25 MilesAnywhere
Across Channel Journey Map
CCBU Experimental
Incubation Series
Orchestration Example- Insurance
<30 seconds wait...Connecting
ŒVisits insurance company online Try get Quote Step 2..4�
“Sarah is looking to switch her car &
home insurance to save money”
“Let’s see if I can save on
my insurance?”
“Hmm..that’s a lot of stu
to enter!”
“Ah...if only I could just
meet with a local agent!”
15% DropoJourney Analytics: 25% DropoJourney Analytics: 75% DropoJourney Analytics:
Local Branch Sta
(Regular Sta/Non-Agent)
Work as a Activity
(Kitchen/Dinning Table or at Oce)
Prospective Customer
Carlos starts his day, logs in to Webex Teams
Sets Presence as “Available”
You 09:35
Hey JourneyBot!
JourneyBot 09:35 AM
Hi Carlos!Would you like receive Virtual Walk Ins ?
30 Minutes
Bot
Summons Journey Bot to accept
Walk-ins for next 30 minutes
You 09:36
Set Message “Maybe I can help!..” JourneyBot!
San Francisco
JourneyBot 09:42 AM
Hi Carlos!Sarah is looking to switch Auto & Home
Oer Walk-in
Bot
Ignore (30s)
JourneyBot 09:46 AM
Incoming Walk-In RequestSarah - Auto+Home - 94108 - 2020 Ford Exp.
Accept Video
Bot
Decline
X...
Sip call Initiated via
Webex Teams Widgets
CJO Triggered
Ž Ready to Give up!
Cu
sto
mer
Sid
e J
ou
rney
On
line
Web
Exp
erie
nce
Œ � Ž
Wfx
Sta
ff S
ide
Jou
rney
Wo
rk a
s an
Ac�
vity
Dro
p o
ff In
tercep�
on
w/ W
ebex W
alkin
“Sarah: Great to be talking
with a human agent, I’ll buy!”
Journey Orchestration - Contact Center Walkin
Jou
rney
Orc
hes
tra�
on
Co
nta
ct C
ente
r W
alk-
in
Website
Quote Step-1 Step-2 Step-3
Contact Center
Claims
Customer Support
100%
65%
15%
20%
10% 15% 25% 75%
50% 25% 15%
35%
Quote Purchase
25%15%
75%
Customer Journey Analytics
On
At more than
and if greater than 30 seconds
then within
Journey Orchestrated Actions
Action #1: Journey Drop-o Interception
Journey
Step-3
Wait Time
Action
Walkin Oer
Agent OerContent
+
Distance
5 Miles
25 MilesAnywhere
+ =
Webex Virtual Walk-inCarlos,
Lic Agent #12XX
1750 Van Ness
Or Type to Chat
Hi Sarah, Maybe I can help!,
We are at 1750 Van ness!
Click to talk on video!
WFH Sta Journey
Customer Journey
Full-Time
Employees
21K
Customer Facing
InStore/Branch
9450 (45%)
FT
Contact Center
1370 (7%)
Digital
Just-in-Time
Frictionless
In-Person
+10x Webex Walk-InFull-Time
Employees
21K
Customer Facing
InStore/Branch
9450 (45%)
FT
Contact Center
1370 (7%)
+10x
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
Customer Journey By Example: Online Store
<script async
src="https://cjaas.webex.com/cj.js"></script>
Cjtag('cong', 'PAGE_TRACK', true);
</html>
Cjtag('cong', 'DIV_CONVERSION', 'conversion');
..
Cjtag('cong', 'MEMBERID', ‘12XXX’);
..
<html>
Cjtag('cong', 'DIV_INTERACTION', 'promobtn’');
Cjtag('XX1232'); <- Customer Journey Token
Œ
�
Ž
Cut & Paste JS for Web or Add to Tag Manager
Journey script is aware of the website structure by setting
DIV pattern used on items that are tracking interaction
(clicks, navigation) or conversion (oer display)”
“Automatic Tracking!
Click Throughs on Oer Tracked
<A href="#" class="promobtn red-outline">Shop
now</a>
Key Events Posted via One Line JS
IVR: Enter your member id
A: 12XXX
Connected
Agent:14XX
Now
Speaking!
2m ago
Added to Cart: Advil 250mg
3m ago
$15 Oer Clicked
4n ago
Member Logged In
3 Days Ago
Checkout, Value $72
Cjtag('post' { 'Addded to cart', 'Advil 250mg' });
JS One-liner:
Agent Dashboard
Emails Chats
CCBU Experimental
Incubation Series
”
Around the World in 80 EventsThe “Tape” (~Events Ledger) Illustrated:
Example Journey Events Recorded on Tape
“Proles” (~View) Illustrated:
Proles are simply curated "View/query" of the Tape with selected attributes and not a full data collection itself. Views are cached for performance and updated upon access (ex: all hits within
30 min may have the same snapshot), Identity resolution is external (ex: Human could merge two IDs together, after which all new events will go to newest prole). There can be more than one
"prole" view/query customized to industry/segment/department or on function (AI Bot using for routing). For instance, a “basic prole” showing just customer name, city and product is
appropriate for Level-1 support, while billing support can use a view that includes home address, credit score, invoice amounts. Views are built always by scanning the Tape from latest to
oldest.
First
Tim
e C
alle
rK
now
n C
alle
rA
ltern
ate
Tim
elin
es
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }{ product: insurance }
Id: xxx-yyyy-zzz
Event: Navigated
IVR Navigation to Insurance
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Connected to AI Bot Agent
For Service
Establish ID (Member #
123XX) via CRM Dip Or Fresh
Record Create
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
Event: Connected
{ Agent: BOTAI } { phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }{ Name: Jane.., Id:123XX }
Event: Prole
Id: xxx-yyyy-zzz^
Escalate to Human Agent-1
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: 123XX
Event: Connected
{ Agent: MIKE }
Call Wrap w/ Resolution &
Sentiment
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: 123XX
Event: Call Wrap
{Transcript: .., Res: Quote}{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
Id: xxx-yyyy-zzz*
Event: Incoming Call
{ phone: 555-555-1234 }
Unknown Caller ( No existing
prole found )
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
Id: 123XX
Event: Incoming Call
{ phone: 555-555-1234 }
555-555-1234 = 123XX
On Call Lookup Prole via
ANI/Caller ID (Found)
Request override routing to direct Agent, via Orchestration (AI ensemble or Flow Control).
Connect to Human Agent-2
Free in < 60 seconds, 92%
Sales Close Rate
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }{ Agent: Sarah }
Id: 123XX
Event: Connected
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
If last res = quote & time < 1D, Sort
Agents by Sales Closing Rates &
wait time < 1 minute
Journey Orchestrated:
Call Wrap w/ Resolution &
Sentiment
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{Transcript: .., Res: Sold}
Id: 123XX
Event: Call Wrap
Email Open
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: 123XX
Event: Email Open
{Quote: Auto Insurance}
Re-establish ID (Member #
123XX) via CRM dip
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: 123XX
Event: Prole
{ Name: Jane.., SFO }
Authentication - Fresh Identity (delink ANI from previous profile)
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Event: Prole
Id: 123XX
{ phone: (555-555-1234)}
Establish ID (Member #
134XX) via CRM
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Name: Jake.., SLT }
Id: 134XX
Event: Prole
Phone Number Recycled
(or Added)
Anonymous Identify Authenticated / Existing Customer
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }{ City: SFO, Browser: FF }
Id: 100.212.21.10* (IP)
Event: Homepage
Anonymous website visitor
arrives on site(or app)
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }{ Page-Title: Auto }
Id: 100.212.21.10 (IP)
Event: Click
Navigate to Auto
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: 100.212.21.10 (IP)
Event: Hover
{ product: EV/SUV }
Hover On EV/SUV
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
Id: 100.212.21.10 (IP)
Event: Request Quote
{ Car: Ford E, Zip: 94102 }
Request Quote Step-1
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: 100.212.21.10 (IP)
Event: Request Quote
{ Name: Jane, VIN: 122Z }
Request Quote Step-2
Rule set by Manager or AI
Idle for 2 minutes on step-2 (Drop off predicted at 75%)
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Id: xxx-yyyy-zzz
details:
{ Tansfer: AIAgent-1 }
Id: xxx-yyyy-zzz
details:
{ phone: 555-555-1234 }
Journey Orchestrated:
If last step = 2 & time > 1 min, Try
Oer Webex Walkin, Within 25
miles
Webex Virtual Walk-inCarlos,
Lic Agent #12XX
1750 Van Ness
Or Type to Chat
Hi Sarah, Maybe I can
help!
Onlin
e
Tape is the collection of journey points that is posted as a series of events from one or multiple sources (web, contact center, mobile app, crm). Tape is an immutable (i.e read-only) running ledger
of the highest granularity of data collected, while proles are higher level views selectively aggregating data from the running Tape.
Display Event/Attrib How Many How Long Type
Name Profile/Name 1 365d Value
City Profile/City 2 365d Value
Phone Profile/Phone 3 365d Value
Items in Cart Add To Cart/SKU - 7d Count
Products Purchased/Product 3 90d Value
Interested Click/Page-Title 1 1d Value
Calls Call Wrap/* 4 30d Value
Emails Email Received/Text 4 30d Value
Chats Chat Wrap/Transcript 4 30d Value
Activity Login - 30d Sum
You name it eld
Value Posted on Event Select this many events latest
Lookback period
Aggregation Type
Example Prole View Built by Business CustomerQuery for “Prole” of 123XX Over The Right Tape
{ City: SFO, Browser: FF }
Id: 100.212.21.10 (IP)
Event: Homepage
Oldest Event
Name:
City:
Phone:
Items In Cart: 4
Activity: 5 (1 Month)
NPS: 7
Products: Sneaker, Bike Interested: Instapot
Jane Doe
San Francisco
555-555-1234 7
ŒŒ �� ŽŽ�
�
�
�‘
‘’
’
Emails Calls
Hello, I Needed if this
product is available in ...
OPEN
RESOLVED 8
Who provides installation
for this product ?
7
RESOLVED 8
7
“
“”
”
9d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
{ City: SFO, Browser: FF }
Id: 100.212.21.10 (IP)
Event: Homepage
9d
{ City: SFO, Browser: FF }
Id: 100.212.21.10 (IP)
Event: Homepage
9d
{ City: SFO, Browser: FF }
Id: 123XX
Event: Homepage
9d
{ Item: Bike }
Id: 123XX
Event: Add To Cart
9d
Event: Prole
Id: 123XX
{ Name: Jane.., SFO }
9d
Id: 123XX
Event: Order Complete
{ Item: Bike }
7d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
Id: 123XX
Event: Email Received
{ Body: Hello I Needed..
7d
{ Rating: 7 }
Id: 123XX
Event: Survey Rating(NPS)
5d
Id: 123XX
Event: Incoming Call
{ phone: 555-555-1234 }
3d
Id: 123XX
Event: Call Wrap
{Transcript:.., Res: Return}
3d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
{ City: SFO, Browser: FF }
Id: 100.212.21.10 (IP)
Event: Homepage
9d
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9d
{ Item: Headphone }
Id: 123XX
Event: Add To Cart
1d
{ }
Id: 123XX
Event: Login
1d
Id: 123XX
Event: Prole
{ Name: Jane.., SFO }
4h
{ Item: Fitness Tracker }
Id: 123XX
Event: Add To Cart
1h
Id: 100.212.21.10 (IP)
Event: Homepage
{ City: SFO, Browser: FF }
9dId: 123XX
Event: Click
{ Page-Title: Instapot }
1h
*Any session id
*Any session id ^Session swapped with Customer ID
Id: 123XX
Ÿ Simpler views help reduce cognitive load on
contact center agents while increasing their
eciency.
Ÿ Oers highest grade of security and data
governance for privacy with
compartmentalization of information queries.
Ÿ Since each category solves a dierent problem
or appeals to dierent functions within a
business, views tailor usability with high
performance lookups.
Retail Tape
(i.e Events Collected)
Œ
Œ
�
Ž
��
�
‘
’“
�
‘
’
“
Originally collected as IP,
but upon Login/Auth
attributed to CustomerID
CRM Dip/Validation
being posted
”
Scan for Name/City
ends right here (latest
Dip Data)
�
�
Originally collected as UUID,
but upon Login/Auth
attributed to CustomerID
Unique Aspects
We show it
Œ
�
Ž
Ž
CCBU Experimental
Incubation Series
© 2021 Cisco and/or its aliates. All rights reserved.
Conclusion
This paper showcases the immense possibili�es of capitalizing Journeys within the contact center - designed to be delivered as part of an CPaaS offering, especially in the post-covid era where social distancing is the new normal. It outlines the innova�ve idea of virtual walk-ins that leverage a companies’ exis�ng larger customer-facing workforce to deliver excep�onal experiences digitally - just-in-�me and fric�onless.
It also a�empts lend common defini�on to customer journeys as it applies to various segments such as Martech, Adtech, CEM & Contact Center.
White paperCisco public
Summary of Orchestration Actions by Value:
Oer/Display Content Oer AI Assist Oer Human Agent Oer Webex Walkin
AI AgentWebex Virtual Walk-in
Carlos,
Lic Agent #12XX
1750 Van Ness
Or Type to Chat
Hi Sarah, Maybe I
can help!
Lower Value Higher Value
Ÿ Browser Intercept*
Ÿ Mobile Push
Ÿ SMS Text
Ÿ Email
Ÿ Browser Intercept*
Ÿ Mobile Push
Ÿ SMS Text
Ÿ Browser Intercept*
Ÿ Mobile Push
Ÿ SMS Text
Ÿ Browser Intercept*
Channel
*When using CJaaS JS Tag
CCBU Experimental
Incubation Series